Exposure to genotoxic carcinogens via both external and internal environments increases risks for major human cancers. Such chemicals form covalent adducts with DNA, which are thought to create a complex program of genetic change acting as the initiating event in malignant transformation and contributing to subsequent tumor progression. Adducts can also disrupt formation or maintenance of normal genome modifications, altering gene expression in affected tissues. Factors that influence formation or repair of adducts are thus likely to be important determinants of human susceptibility to environmental carcinogenesis. We propose an approach to elucidate types of DNA damage and cellular processes responsible for genetic changes that transform normal cells to malignant ones. As a tool, our work focuses on aflatoxin B1 (AFB1), an established risk factor for human hepatocellular carcinoma that strongly elevates risk in synergy with hepatitis B virus infection. As in the human HCC incidence pattern, AFB1 is more potent in males than females, and animals of both sexes are more sensitive as juveniles than as adults. The hypothesis underlying our proposed work is that different quantitative features of mutagenesis (mutation frequencies) or qualitative features (mutational patterns) are determinants of the initiation and promotion phases of tum origenesis. The goal of our work is to characterize these quantitative and qualitative features and define their roles in modulating liver carcinogenesis. Experiments designed to test this hypothesis will employ a newly developed Duplex Sequencing protocol that enables the application of Next-generation sequencing platforms for mutational analysis. Mutation frequencies and spectra will be determined in selected genomic DNAs at stages throughout the tumorigenic process in a mouse model of AFB1-induced liver cancer. These analyses will enable us to determine whether a mutator phenotype is acquired during tumor development induced by AFB1, while providing a picture of its variation among different genes. Evidence of mutator phenotypes is seen in many advanced human tumors, and is increasingly considered as a possible source of premature-onset drug or radiation resistance. The types of mutations that we observe will provide hallmarks of the cellular processes that orchestrate the genetic changes during tumorigenesis. Our animal model is well suited to study the time of onset of a mutator phenotype and, of equal importance, for additional studies to investigate interventions that could delay tumor development and drug resistance.